高斯滤波稳健性能的研究与改进

Study and Improvement of Robust Performance of Gaussian Filtering

以表面轮廓信号为例 ,结合图形和参数评估 ,研究了异常信号对高斯滤波性能的影响。引入稳健统计学的思想 ,对高斯滤波进行了稳健处理 ,并比较了几种稳健估计权函数对高斯滤波性能的改进。结果表明 。

Taking the surface profile signal as an example,it analyzes the influence of outliers on the performance of Gaussian filtering with the help of the illustrations and evaluation parameters.Robust statistics is introduced to enhance the robustness of Gaussian filtering.Several selected robust weight functions are adopted and compared.The results show the presented robust Gaussian filtering is effective to eliminate the influence of outliers and make Gaussian filtering more robust.